A data network is conceivable where numerous sensors, for example, several hundred to several tens of thousands of low-cost chip sensors (data processing apparatuses) are installed in an installation area and each sensor wirelessly transmits detection related information, and a higher level parent device collects the detection related information. When implementing wide area coverage using numerous sensors, sensors have to be provided in an enormous number (nodes).
Further, sensors that operate using a harvester (e.g., energy conversion of light, vibration, temperature, wireless radio waves, etc. in the external environment of the installation site) as a power source, have extremely low processing performance. When a data network includes non-guaranteed sensors, the non-guaranteed sensors have to be removed from the system to guarantee the accuracy of data. However, each sensor is of an extremely small size and is difficult to visually identify and therefore, cannot be easily removed.
In a typical server client system, client authentication is performed using encryption that employs, for example, public keys and private keys, and written certification in which these are used. When a sensor performs processing to execute an authentication scheme, both the volume of data and the volume of processing create an extremely large processing load and therefore, with respect to simple sensors, there is a need for a reduced volume of data that is to be handled and authentication that can be performed easily.
According to a conventional technique, in a sensor network, sensors that perform detection according to a sensor attribute (temperature, humidity, position, etc.) are identified from among plural sensors (for example, refer to Japanese Laid-Open Patent Publication No. 2008-176502). According to another technique, nodes are identified using ID information that is transmitted from a system to the nodes to control the power source of each node (for example, refer to Japanese Laid-Open Patent Publication No. 2010-141469).
With the conventional techniques above, no measure to guarantee of the accuracy of the data output by the sensors is performed. For example, whether the large number of installed sensors performed a proper operation test is not confirmed. In this case, if sensors that have not performed a given authentication test (operation test), such as sensors of an unauthorized manufacturer, are installed in the installation area, the accuracy of the data collected from the sensor information decreases.
Further, sensor reliability such as data accuracy, durability, etc. may vary according to manufacturer. Therefore, after sensor installation, the operation state of the sensor during actual operation cannot be determined and assessment of the sensors cannot be performed. Additionally, when sensors of multiple manufacturers are installed in an installation area, no process can be implemented for managing assessment according to manufacturer and managing distribution of utilization costs for sensor operation according to manufacturer, based on the assessment results.